Gleb Nosovskiy
(also spelled Gleb Nosovsky) is a Russian mathematician and publicist best known as a primary co-author of the New Chronology theory alongside Anatoly Fomenko. Born on January 26, 1958, in Moscow, he is a graduate of Moscow State University and has a background in probability theory and computer geometry.
Key Facts

Role in New Chronology: Since 1995, Nosovskiy has co-authored numerous books with Fomenko, claiming that conventional world history was largely falsified or misdated between the 16th and 19th centuries.
Controversial Claims: His work suggests radical historical revisions, such as dating the life of Jesus to 1152–1185 AD and reimagining a "Rus-Tartary" empire.
Academic Reception: While he holds a degree in physics and mathematics, his historical theories are widely rejected by the mainstream scientific and historical communities as pseudohistory or pseudoscience.
Current Standing: As of 2026, he remains associated with the Moscow State University Department of Mathematics and Mechanics and continues to promote the New Chronology through various multimedia platforms.

Variations in Spelling

Gleb Nosovskiy: Commonly used in the noscripts of his English-translated book series, History: Fiction or Science?.
Gleb Nosovsky: Frequently used in academic citations, library catalogs, and Wikipedia.
Глеб Носовский: The original Russian spelling of his name.

Gleb Nosovskiy’s career is primarily defined by his transition from mainstream mathematics at Moscow State University to the radical historical revisionism known as "
New Chronology."
Timeline of Key Milestones

1958: Born on January 26 in Moscow.
1980s: Attended lectures by Professor Mikhail Postnikov at Moscow State University, which introduced him to the chronological revisionist ideas of Nikolai Morozov.
1995: Formally began his permanent co-authorship with Anatoly Fomenko, marking the start of the "New Chronology" book series.
2003: Published work applying astronomical methods to date the life of Jesus to 1152–1185 AD.
2019: The "Multimedia Museum of a New View on History" opened in Yaroslavl, dedicated to the work of Fomenko and Nosovskiy.
2023–2024: Published latest major English reconstructions, including Volume 8 of the core series.
2026: Continues as a faculty member at Moscow State University in the Department of Mathematics and Mechanics while promoting revised global history.

Integrated Bibliography (Major English Works)
The following is an integrated list of key publications from the History: Fiction or Science? series, which serves as the comprehensive foundation of his work. These books are often released in multiple editions; dates reflect significant English-language versions available on platforms like Amazon or AbeBooks.

Chronology 1: Dating Methods as Offered by Mathematical Statistics (2016/2026 Reprint) – Details the statistical flaws in traditional dating.
Chronology 2: The Dynastic Parallelisms (2018) – Uses mathematical modeling to argue that different historical dynasties are actually "duplicates" of the same rulers.
Chronology 3: Astronomical Methods as Applied to Chronology (2007/2018) – Re-dates the Almagest of Ptolemy and Egyptian zodiacs to the 11th–16th centuries.
Chronology 4: Russia, Britain, Byzantium, Rome (2008) – Claims "Ancient" Rome and Byzantium were medieval entities.
Chronology 5: Russia = Horde (2020) – Identifies the Mongol Empire with a medieval Russian-Tartar state.
Chronology 6: The Horde-Ataman Empire & The Bible (2020) – Reinterprets Biblical history and the conquest of America.
Chronology 7: A Reconstruction of Global History (2024) – Provides a synthesized timeline of the proposed "Great Empire".
Volume 8: Reconstruction of Chronology (September 2023) – Features a "fully improved, cross-checked bibliography" and summarizes 40 years of research.

AI Overview
+7
Dark Web searches differ from conventional ones because
they access intentionally hidden, unindexed content requiring special browsers (like Tor) and unique .onion URLs, focusing on anonymity via layered encryption (onion routing) rather than public indexing, making them decentralized, less reliable, and often requiring direct links or dark web-specific search engines (like Torch) to find content that is intentionally difficult for standard crawlers to find.
Conventional Search (Surface Web)

Indexing: Uses massive crawlers (spiders) to map and index billions of public web pages, making them searchable via keywords.
Access: Standard browsers (Chrome, Firefox) and typical URLs (e.g., .com, .org).
Anonymity: Low; IP addresses and browsing are traceable.
Structure: Centralized search engines (Google, Bing) organize results.

Dark Web Search

Indexing: Actively avoids standard indexing; relies on specialized dark web search engines or direct links.
Access: Requires anonymizing software like the Tor browser and specific ".onion" addresses (e.g., eajwlvm3z2lcca76.onion).
Anonymity: High; uses "onion routing" to obscure IP addresses through multiple relays.
Structure: Decentralized, with content frequently appearing/disappearing; relies on community input, forums, and manually shared links.

Key Differences Summarized

Visibility: Surface web is indexed; Dark Web is hidden.
Tools: Surface uses normal browsers; Dark Web uses Tor.
Addresses: Surface uses standard URLs; Dark Web uses complex .onion links.
Reliability: Surface is generally stable; Dark Web content changes rapidly.
Purpose: Surface for general info; Dark Web for anonymity (legal and illegal).


Dive Deeper in AI Mode

"Dark Web" searches in 2026 differ fundamentally from conventional surface web searches (like Google or Bing) in their accessibility, indexing methods, and underlying technology.

Key differences include:
1. Accessibility and Specialized Software

Restricted Access: Conventional searches are performed via standard browsers (Chrome, Safari, Firefox) on the "surface web".
Specialized Browsers: Dark web content is hosted on encrypted overlay networks, or "darknets," primarily the Tor network. Accessing these sites requires specialized software like the Tor Browser, which handles the multi-layered "onion routing" needed to reach .onion domains.

2. Indexing and Search Scope

Active Resistance: Unlike the surface web, which is designed for discovery, the dark web is built for anonymity and often actively resists crawling and indexing.
Unlinked Infrastructure: Conventional search engines use automated "crawlers" that follow links from one site to another to map the web. Dark web sites are often not interlinked, making traditional crawling largely ineffective.
Manual Submissions: Dark web search engines (e.g., Ahmia or Torch) frequently rely on manual user submissions or specialized Tor-based crawlers that are slower and less comprehensive.

3. Ranking and Result Quality

Basic Algorithms: Surface web engines use complex algorithms based on authority, relevance, and user behavior to rank results. Dark web engines typically use much simpler ranking systems, often prioritizing site recency or manual curation.
Volatility: Dark web sites are notoriously transient; they often go offline or change addresses frequently to avoid detection, leading to many "broken" or outdated links in search results.

4. Content and Filtering

Unfiltered Results: While Google blacklists or deprioritizes harmful content, many dark web search engines like Torch or Haystak are uncensored, displaying raw results that may include illicit marketplaces, scams, or malware.
Safety-First Alternatives: Some specialized engines, such as Ahmia, voluntarily filter out illegal materials (like child exploitation content) to provide a "cleaner" experience.

1. The Automation Toolkit (Script & Shell)
For bulk processing, developers generally avoid "macros" (which simulate clicks) in favor of APIs and headless noscripts.

Primary Script: Python is the industry standard. Packages like ytfetcher or youtube-trannoscript-api are used to gather trannoscripts from entire channels or playlists in minutes.
The Shell: These noscripts are typically run in a Bash or Zsh environment (Linux/macOS).
Best Distro: Ubuntu or Debian are the most common Linux distributions for these tasks due to the wide availability of Python libraries and automation tools.
The Process:
Fetch all video IDs from a channel using the YouTube Data API.
Use an asynchronous Python noscript (like yt-fts) to download all available trannoscripts as JSON or text files.
Ingest these files into a local search engine like Elasticsearch or Meilisearch to enable instant, "pithy" context searching across the entire archive.

2. Search Services & Commercial Vendors
Several companies sell specialized "Conversation Intelligence" or "Video Scraping" services that pre-index content or provide high-speed batch processing.

Reduct.Video: Best for researchers and lawyers needing to triage thousands of hours of footage with 99% accuracy.
Sonix.ai: High-accuracy multilingual trannoscription with a robust API for bulk automation.
Apify / Oxylabs: These vendors sell Scraping APIs that bypass rate limits to extract metadata and trannoscripts at scale from platforms like YouTube and Rumble.
Praiz.io: Focuses on turning large volumes of video calls into searchable, structured data for sales teams.

3. "Dark Web" and Custom Indexing Logic
Regarding your question on building indexes from scratch:

Building the Index: Search engines (on the clear or dark web) use Web Crawlers (or "spiders") to methodically browse the web and follow links to discover new content.
Archiving: New indices are often stored in distributed vector databases (like Pinecone or Milvus in 2026). This allows the engine to perform "semantic searches," finding the meaning of a spoken phrase rather than just exact keywords.
Iterating Queries: To keep an index current, "delta crawlers" are used—noscripts that only look for changes or new uploads since the last crawl timestamp. Older indices are moved to "cold storage" (archived) to save on performance costs while remaining accessible for historical research.

4. Capabilities in 2026
As of January 2026, I can:

Identify the exact Python libraries and APIs required to build your own bulk searcher.
Analyze text-based trannoscripts from indexed videos to find specific mentions or context.
Recommend specific commercial vendors that handle the heavy lifting of rate-limiting and high-speed data extraction.

Yes, it is possible to search dialogue from many videos on YouTube and other platforms using current technology, but it often requires third-party tools to search across an entire library. This technology is already accessible and widely used.
Searching YouTube and Social Media Dialogue

Within a single YouTube video: You can use the built-in trannoscript feature. Click the "More" tab (three dots) below the video and select "Show trannoscript". A panel with the full, timestamped text will appear next to the video, which you can then search using your browser's find function (Ctrl+F on Windows or Cmd+F on Mac).
Across multiple videos/channels: Standard YouTube search is keyword-based and not designed for deep dialogue search across an entire channel's backlog. However, several third-party AI tools and services are currently available that allow you to upload video files or paste URLs and then perform keyword or even semantic searches across all the resulting trannoscripts in one place.
Examples include: Filmot, which searches YouTube captions; specialized AI platforms like Imaginario AI or Sonix that create searchable libraries of your content; and Python noscripts like yt-fts for power users.
Other Platforms (Rumble, etc.): The approach is similar. You would typically need to use a third-party trannoscription service (like Rev, Otter.ai, or Veed) to generate a searchable text trannoscript of the videos.

Accessibility and Development

Accessibility: This technology is already accessible to the public. Basic trannoscription search on individual YouTube videos is a built-in feature, and numerous commercial AI-powered video search and trannoscription tools are available for free or as paid services right now.
Dark Web: There is no indication that such tools are a specific development on the "dark web." The technology is mainstream, developed by legitimate AI companies and platforms to make video content more accessible, searchable, and useful for content creators, researchers, and general users.

These articles explain how to use AI trannoscription tools to search video dialogue and discuss the current accessibility of this technology:

Automating the search of video dialogue is already a standard practice in 2026, using a combination of specialized AI tools and browser automation.
Methods of Automation

Browser Automation & Scripts: You can use tools like Selenium or Puppeteer to create noscripts that simulate a human opening a video, clicking "Show Trannoscript," and scraping the text. For less technical users, browser extensions like Tampermonkey allow you to run custom JavaScript (UserScripts) that can automatically expand and search trannoscripts across multiple tabs.
Low-Code/No-Code Workflows: Platforms like Zapier and Make (formerly Integromat) allow you to set up "Triggers." For example, a "YouTube Trigger" can detect a new video, send the link to an AI like ChatGPT for trannoscription/analysis, and then log specific keywords into a Google Sheet.
Specialized AI Tools: Services like Sonix, HappyScribe, and VOMO AI are designed specifically for high-accuracy bulk trannoscription. These tools often provide APIs that allow you to integrate "dialogue search" directly into your own custom software or databases.

Search Engine Pre-Indexing
Search engines have evolved to move beyond text-only pages:

Major search engines, like Google and Bing, now use "multimodal" indexing. This means they index trannoscripts, audio patterns, and video scenes.
Google's Gemini-powered Deep Dives and tools such as Perplexity allow natural language "visual searches" and dialogue extraction directly from search results. This eliminates the need to watch the video first.
Emerging engines, including Sigma Chat and Bagoodex, specialize in using web video dialogue to answer user queries conversationally.

Search Capabilities
Searches can be performed in the following ways:

Real-time web analysis can search and synthesize information from indexed trannoscripts in search results or specialized databases.
Multiple distinct searches can be executed simultaneously to cross-reference data across different platforms, such as YouTube and social media.
Analysis of large volumes of text-based video data can find specific claims, quotes, or recurring themes.
Specific noscripts or third-party tools required to perform "offline" or "deep" searches can be identified.

In typical holometabolous insects, the
larval stage generally has a higher brain-to-body mass ratio than the adult stage because most body mass accumulates during the last larval instar while brain cell number increases less dramatically, or brain growth is complete earlier. In contrast, in species like Drosophila, the adult brain has a significantly higher number of neurons (40 times more) to support complex adult behaviors.
Larval vs. Adult Brain-to-Body Mass Ratio

Larval Stage: The larval brain is a simpler structure with fewer neurons, adapted for functions like feeding and growth. Body mass increases exponentially, especially during the final instars, while the total number of brain cells increases asymptotically or at a slower rate in comparison to body growth. This results in a higher overall body mass relative to brain mass by the end of the larval stage compared to earlier instars or the eventual adult stage.
Adult Stage: The adult brain undergoes significant remodeling and neurogenesis during metamorphosis, developing complex structures like the mushroom bodies and antennal lobes to support complex adult behaviors such as flight, navigation, foraging, and mating. While brain mass also increases, it is typically an allometric relationship where larger animals have proportionally smaller brain-to-body ratios overall (Haller's rule). The specific requirements of the adult lifestyle often lead to an effective increase in "brain power" relative to functional needs, but this doesn't always translate to a raw mass ratio increase over the final larval stage's ratio, due to the massive body size of the final instar larva.

Gender Differences in Brain-to-Body Mass Ratio
Gender differences in brain-to-body mass ratio are often linked to the division of labor and specific behavioral requirements within a species.

In social insects like ants, males often have disproportionately smaller heads and less complex behaviors compared to workers (females), leading to different brain size investment. Worker ants can have a brain that is about a seventh of the insect's total weight, a much higher ratio than most other animals, reflecting their complex tasks.

Strepsitera (Stylopidia) Specifics
The case of Strepsiptera is highly specialized due to the extreme sexual dimorphism and female neoteny.

Males: Free-living adult males have well-developed sensory organs, wings, and legs to locate and mate with females. They require the associated large brain regions (e.g., optic lobes, antennal lobes) to process these complex sensory inputs and perform flight and navigation.
Neotenous Females: Adult females of most Stylopidia species are neotenous, meaning they retain larval characteristics into adulthood. They are endoparasitic, lack sensory organs (compound eyes, antennae), legs, and wings, and remain inside the host. Their nervous system is greatly simplified, reflecting their sessile, non-foraging, and non-navigating lifestyle within the host.

This dramatic difference in physical structure and behavioral needs results in a substantial difference in brain investment. Male Strepsiptera likely have a much larger relative brain mass/volume compared to their body size than the highly simplified, neotenous females, whose neural tissue requirements are minimal to manage basic life functions and reproduction within a host. The female's body is primarily an egg-laying machine (larviform) and does not need a complex, energy-intensive brain